Food insecurity and low diet quality are persistent problems linked with chronic disease and poor health among limited-resource children and adults using Supplemental Nutrition Assistance Program (SNAP). We have shown nutrition education via adult-focused, direct SNAP-Education (SNAP-Ed) improved household food security by 25% but not adult dietary quality among SNAP-eligible households using a randomized, controlled, longitudinal SNAP-Ed intervention in Indiana. Households experiencing food insecurity often reserve food considered "healthful" for children, so child dietary quality improvement may precede that observed among adults when household food security improves. This study will determine the effect of adult-focused direct SNAP-Ed on child dietary quality and household food security using a longitudinal randomized, controlled SNAP-Ed intervention. Assessment will include repeated 24-hour dietary recalls to determine usual intake, the U.S. Household Food Security Survey Module, and behavior data from before and after the 10-week "intervention period," and 1 year later, after which the control group will receive the intervention. Low-income participants (n=275) from Indiana will be recruited following SNAP-Ed protocol. Results of the study will inform the creation of supplementary on-demand SNAP-Ed educational material focused on improving healthful dietary intake for children and adults in situations of food insecurity in households with children. Education on modeling healthy attitudes and behaviors, planning and preparing family meals, and dietary shortfalls as informed by the results and previous evidence will be included and evaluated. The study aligns with the goals of USDA to increase food security and this RFP to improve healthful behaviors, food quality and nutrition.

The goals of SNAP-Ed are to "improve the likelihood that persons eligible for SNAP will make healthy choices… consistent with the current Dietary Guidelines for Americans." Our recent review found four studies evaluating household food security as an outcome of direct adult SNAP-Ed with evidence considered strong because of our studies that used randomized and controlled study designs and assessment using the U.S. Household Food Security Survey Module (HFSSM).We showed that nutrition education via adult-focused, direct SNAP-Ed improved household food security by 25% among SNAP-eligible households over 1 year after a SNAP-Ed intervention in Indiana. However, previous evidence of meaningful change in dietary quality and intake among adult SNAP-Ed recipients is less secure despite demonstrated changes in attitudes, self-efficacy, intentions, and behaviors toward increasing intake of nutrient-dense foods due to a lack of randomized, controlled, longitudinal study designs and inherent measurement error in the assessments quantifying diet. Importantly, our study, using a longitudinal, randomized, controlled design and estimating usual intake with best practice assessment and analysis, did not determine dietary quality change among adults nor other dietary components except for vitamin D. Households experiencing food insecurity often reserve food considered "healthful" for children so child dietary quality improvement may proceed that observed among adults when household food security is improved. As children age, however, diets are known to worsen. Therefore, dietary quality improvement over time due to SNAP-Ed will need to overcome a strong influence of age, yet even maintained dietary quality over time would be a success compared to the expected declines in dietary quality for a control group whose parents do not receive SNAP-Ed. Taken together, we hypothesize that dietary quality among children and household food security will improve 1 year after a direct, adult-focused SNAP-Ed intervention compared with a control group while adult dietary quality will not improve compared with the control group.

Long-Term Goal: Determine the long-term (1-year) effect of direct, adult-focused SNAP-Ed on dietary quality and household food security among children (5-18 years) of adult (≥18 years), low-income Indiana SNAP-Ed participants.

(a) dietary quality (using the HEI), usual dietary intake, among children of adult SNAP-Ed participants and the adult participants themselves, compared with a control group receiving a delayed intervention and (b) food security among the household, adults and children (using the U.S. HFSSM) in intervention compared with control group, and (c) coping mechanisms, needs, behaviors, and attitudes among adult SNAP-Ed participants compared with controls.

1. Objectives Research Objectives

1. Recruit and randomize n=275 SNAP qualifying, low-income adult (≥18 years) and child (5-18 years) pairs from Indiana to participate in this longitudinal study with approximately n=138 pairs receiving the SNAP-Ed intervention and n=138 pairs in the control group (not receiving SNAP-Ed until after the study).
2. Quantify child and adult dietary quality (HEI) and usual intake (2, 24-hour recalls at each time-point), household, adult and child food security, and additional behavioral, coping mechanisms, needs, and self-efficacy at baseline and 1 year after the intervention period.
3. Evaluate change in child and adult dietary quality and usual intake per DGA compliance and food components of public health relevance (as per Dietary Reference Intake standards); household, adult and child food security status; from baseline one year follow-up for intervention compared with control group. Due to nonresponse and loss-to-follow, we expect roughly 100 pairs in each group to provide change measures.

Goal 1:Research Study Participants and Recruitment Recruitment will occur at a rate of n=10 per week for 9 months in the 4 regions of the state (northwest, northeast, southwest, southeast) over the course of the school year so that each region is visited roughly once per month to avoid a seasonal effect. Approved SNAP-Ed sites within regions will serve as recruitment sites including community centers; SNAP, WIC, county Extension offices; food pantries and banks, and other locations. An informational flyer will be posted at the site on study recruitment days. A full-time study recruiter, trained through SNAP-Ed, will coordinate to recruit alongside the NEPAs in the 4 state regions who will be present to similarly recruit for SNAP-Ed programming and deliver the lessons. The recruiter will explain the study procedures, assessments, and time necessary to complete the assessments. The recruiter will gauge interest and administer the screener querying exclusion criteria. The consent form, randomization, and study assessments will follow if the individual desires to participate, if not they will be referred to the NEPA and engage in SNAP-Ed as usual without differential treatment through the program. Assent for children 5-17 years and consent for children 18 years and assessments will be scheduled with the recruiter within the following 24 hours by phone or by email when the participant desires. Participants joining the study will be queried as to whether they prefer assessment via interview or independently online and similarly their preference for completing the 1-year assessments. The reason for any non-response will be tracked if known. Participants without email, without access to a computer with internet, or with limited computer skills may also complete the assessments via telephone with assistance from researchers. Treatment group (control or SNAP-Ed intervention) will be assigned 1:1 using data collection spreadsheets that will include the subject ID and randomized assignment. Participants recruited simultaneously will be assigned to the same study group to prevent knowledge of differential treatment. Subject ID will be assigned to each assessment the participant completes; names will not be associated with any assessments, and the list of names and numbers will be kept separately and securely. A password protected metadata file on an encrypted hard drive or REDCap or Box will securely store all information.

Research Study Protocols: Adult participants will complete assessments following recruitment. The participant screener, characteristics and additional behavioral, coping mechanisms, needs, self-efficacy, and food security questionnaires will be completed online or interview assisted in Qualtrics or REDCap. Child participants will complete the child-specific questions of the U.S. HFSSM, and a brief set of behavioral, and coping mechanisms questionnaires online (with interview and parent assistance for ages 6-11 years and with interview and option to self-complete for ages 12-18 years). Study assessments for the intervention group will be provided alongside the SNAP-Ed program evaluation before the first lesson. The 24-hour dietary recall will be self-completed online or via interview for adults using the Automated Self-Administered 24-hour (ASA-24) Dietary Recall and by interview with parent assistance ages 6-11 years and by interview or self-completed for ages 12-18 years through a secure data management system. The same sequence of assessments will be completed at each time point, with the exception of the participant screener completed at baseline. Assessments will include: (1) participant screener, characteristics, behavioral, coping mechanisms, needs, self-efficacy, and U.S. HFSSM using a 1-year reference period and (2) SNAP-Ed curricula program evaluation, (3) recruitment day dietary assessment using a 24-hour dietary recall for adult and child, and (4) non-consecutive day follow-up dietary recall for adult and child. The characteristics portion of the survey will query personal and household characteristics.

After baseline study assessments are completed, participants randomized and recruited to the intervention group will be referred to the SNAP-Ed NEPA and receive the first SNAP-Ed lesson. The remaining 3 lessons will be received at the rate of 1 lesson per week during the 10 week intervention period, following normal protocol. Control group participants will not receive lessons during the intervention period and will be requested to wait 1 year to receive lessons. All recruited participants (control and intervention groups) will be contacted at least once per month via communication of their choice to reduce attrition and maintain correct contact information until completion of the final survey. SNAP-Ed lessons will be delivered either one-on-one or in a group setting at a community location, following normal protocol.

The study recruiter will follow-up with NEPAs delivering lessons to intervention participants to track lesson attendance, location, NEPA delivering the lessons, delivery to individual or group, contacts with the NEPA, and presence and type of county-level indirect SNAP-Ed for each participant using a spreadsheet in Qualtrics or RedCAP. Next, the study recruiter will request to keep in touch with each participant to update contact information on a monthly basis to minimize attrition. After 1-year, participants will complete 1-year assessments and the control-group referred to SNAP-Ed.

Success of the randomization for the Goal 1 will be checked by comparison of the experimental group with the control group for each measured independent variable for all characteristics variables and potential confounders. The interaction of time and treatment group in this model will be the main independent variable of interest while the various outcomes will be the main dependent variables in different models. The difference of differences will be determined by comparing changes in outcomes from baseline to the 1-year follow-up in each treatment group using linear contrasts. The model conditions necessary for inference will be checked through residual plots, Q-Q plots, and histograms. Remedial measures such as transformations, will be completed to address non-normal distributions. Least squares mean estimates of the 1-year change by treatment group and their difference (i.e., the interaction) will be reported along with their standard errors. Results will be significant when p≤0.05 performed using SAS.

Dietary quality and intake: Adult and child participants with at least one complete ASA-24 recall and complete baseline survey information will be included in the analysis of diet quality and intake. Outcomes will be the difference of the change in usual dietary quality and component scores and usual intakes by time between the intervention and the control group for nutrients and food groups of interest, and the proportions of the study group not meeting the DGA, EAR, or exceeding the AI. To assess overall diet quality, the HEI-2015 scores will be analyzed between groups over time. The NCI method will be used to estimate distributions of usual intakes and the proportion of the population at risk for nutrient inadequacy based on the cut-point method and probability method for iron. Models will be adjusted for covariates treated as fixed effects: age, energy as kilocalories, and variables indicating the day of the week of the recall (weekday (Monday-Thursday) or weekend day (Friday-Sunday)) and the sequence of the recall (1st or 2nd). Other covariates described for food security models will also be considered.

Inclusion Criteria:

• households in Indiana
• households with children
• English speaking
• eligible to receive SNAP (≥18 years and household income at or below 130% of the poverty guideline)
• willing to allow a child 5-18 years to participate
• willing to participate in the study and wait 1 year to receive SNAP-Ed

Exclusion Criteria:

• not have received SNAP-Ed lessons in the past year
• not pregnant or lactating (due to inherent dietary changes)